Effect of undetectable PSA following SRT for biochemical recurrence on BPFS, FFM, and OS.

367 Background: Salvage radiotherapy (SRT) alone or in combination with androgen deprivation therapy (ADT) is the only curative treatment option for patients with biochemical recurrence after radical prostatectomy (RP) for prostate cancer. An undetectable post SRT PSA in patients without ADT (<0.1 ng/ml) has been shown retrospectively to correlate significantly with long-term freedom from progression, metastasis free survival and overall survival. The aim of this study was to investigate whether the predictive power of a post SRT-nadir <0.1 ng/ml extends to all subgroups of SRT patients. Methods: Between 1998 and 2018, 698 patients (all pN0/cN0) with persisting PSA or increasing PSA out of the undetectable range (<0.1 ng/ ml) of two German university hospitals received SRT alone without ADT. Exclusion criteria were distant metastases and/or ADT in the interval between RP and/or in combination with SRT. ADT before RP was allowed. All patients received 3D-conformal or intensity-modulated SRT to a median dose of 70.2 Gy. The impact of post-SRT nadir <0.1 ng/ml on biochemical progression free survival (BPFS), freedom from metastases (FFM) and on overall survival (OS) was analyzed by Kaplan-Meier analysis and multivariate Cox regression analysis. 9 subgroups with at least 30 patients per group were analyzed separately (pT2 + R0, pT3 + R0, pT3 + R1, pT2 + Gleason Score ≤ 7, pT3 + Gleason Score ≤ 7, pT3 + Gleason Score ≥ 8, pT3 + Gleason Score ≥ 7 + R1, pT3 + Gleason Score ≥ 7 + pre-SRT-PSA-Wert ≤1,0 ng/ ml and pT2/3 + Gleason Score ≥ 8 + pre-SRT-PSA-Wert ≤1,0 ng/ ml). Results: Median follow-up was 5.6 years. Median PSA before SRT was 0.30 ng/ ml. A post SRT PSA <0.1ng/ml was achieved by 76.6% of patients. In univariate analysis of the entire cohort, post SRT PSA <0.1 ng/ ml was a significant predictor of BPFS (p<0.001), FFM (p<0.001) and OS (p=0.01). Regarding univariate analysis of the subgroups, post SRT PSA <0.1ng/ml was a significant predictor of BPFS in all subgroups (p<0.001), FFM in 6 subgroups (p<0.001 to p=0.03) and for OS in 3 subgroups (p<0.001 to p=0.02). In multivariate analysis of the entire cohort, post-SRT PSA <0.1ng/ml was an independent predictor for BPFS (HR: 7.25, 95%-CI: 5.36-9.81, p<0.001), FFM (HR: 4.21, 95%-CI: 2.00-8.84, p=0.002) and OS (HR: 2.59, 95%-CI: 1.28-5.23, p=0.008). Conclusions: Achieving an undetectable PSA after SRT is an independent predictor of BPFS, FFM and OS. This effect is present in subgroups with favorable (pT2+Gleason score ≤ 7) as well as in those with unfavorable characteristics (pT3+ Gleason score >8). Our findings indicate that ADT could be withheld in a proportion of patients that are candidates for ADT based on their risk factors but achieve an undetectable PSA after SRT. Prospective studies are needed to confirm these results.